Silhouette Loss: Differentiable Global Structure Learning for Deep Representations

arXiv:2604.08573v1 Announce Type: cross Abstract: Learning discriminative representations is a central goal of supervised deep learning. While cross-entropy (CE) remains the dominant objective for classification, it does not explicitly enforce desirable geometric properties in the embedding space, such as intra-class compactness and inter-class separation. Existing metric learning approaches, including supervised contrastive learning (SupCon) and proxy-based methods, address this limitation by o